Andy Bolton, Chief Executive Officer, Capacitas

Professor Paul Hardaker, Chief Executive, Royal Meteorological Society

ABSTRACT: Making forecasts of the weather requires some of the world's leading edge technology, in data communications, super-computing facilities and in the applications of e-Science (Web Services technology), and often the science is waiting for the technology to catch up. But new developments in forecasting at small scales and for longer durations is stretching our knowledge of the science, as much as when the meteorological community introduced the idea of numerical weather prediction (NWP).

In the UK we use a single Unified Model for our weather forecasting and for our climate predictions, where much of our knowledge of the physics and dynamics of the atmosphere is brought together with complex ocean models to provide our predictions of future climate. The UK's climate predications are respected around the world and were a major contribution to the Inter-Governmental Panel for Climate Change (IPCC) 4th Assessment Report (4AR). The 4AR showed some major developments in our knowledge of our changing climate and in early 2009 the UK's new climate scenarios will be published, bringing for the first time probabilistic information on climate for the UK business community.

In the next 5 years the UK meteorological community faces some interesting and difficult choices about the future direction of its weather and climate modelling and will need new additional and significant funding to support the development of the next generation of its prediction models.

ABSTRACT: The following quantitative claims are made in the marketing materials of Server Virtualisation Vendors

• Most servers and desktops today are in use only 8-15% of the time they are powered on
• Consolidation ratios of 15:1 can be achieved
• Increase hardware utilization to as much as 85%
• Reduce energy costs and consumption by 80%

ABSTRACT: Green computing is now high on the agenda of many CIOs and may well become a legal requirement soon. Good Capacity Management is essential to the efficient running of the datacentre which will ensure that costs are kept to a minimum. Using these same techniques it is also possible to make your datacentre Green. This presentation explores how good Capacity Management practices can be employed to get the most from your IT systems whilst ensuring the minimum of CO2 is produced thus improving the Green credentials of your organisation.

ABSTRACT: Many organisations have performance test environments for key systems. It is often thought that a performance test environment should be similar in size to the production system in order to get representative performance metrics. The cost of a production size performance test environment can be prohibitively expensive. In the past, the general view was that IT hardware was cheap therefore having a large performance test environment was deemed to be possible. However, this is no longer the case, due to the rising cost of data centres, e.g. power/space etc. Typically, data centre hosting costs for a single server are £12k per annum. Software license costs can also add a significant amount to this bill.

The requirement for a production size test environment may not be essential to support the performance testing activity. This presentation gives an example of the potential cost savings that can be made by using a scaled down performance test environment in conjunction with capacity/performance modelling techniques. The results show that costs savings for a single environment can run into hundreds of thousands of pounds per annum. This presentation also outlines some of the technical considerations that need to be taken into account when using this approach.

ABSTRACT: Data centre costs represent a significant portion of a large organisation’s IT budget. A recent study by the Uptime Institute revealed that 80% of annual IT costs are driven by amortised data centre capital costs. There are significant data centre cost pressures on IT departments resulting from growing demand for server space, spiralling power costs and the rising power density of IT equipment. This presentation describes how capacity planning can be employed to achieve a balance between service level and cost in the data centre. This presentation looks at a number of approaches that can be used to enhance cost control and promote forward cost planning of large data centres.

ABSTRACT: The IBM mainframe is ideally suited to running multiple applications on one or more copies of the same operating system. Since these machines support so many business applications and users the running costs can be high. The complexity of these systems means that support staff tend to focus on either keeping things running or enabling new applications to go into production. This paper shows how a chargeback mechanism can be created by building on an existing capacity management process and how this mechanism can drive tuning and efficiency efforts in order to drive down costs both for specific user groups and for the installation overall.